The Application Of Data Mining Technology In Wireless Location

With the developing of technology,the improving of living standard and the coming of the era of data,the demand of service industry increased greatly.The demand contains Weirless Location Service.Therefore,various location techniques have obtained fast development.Some are familiar with us,such as Baidu Map,Google Map,Amap and so on.They all have great mature location technologies.Base Station Location Technology has rapidly progress in the mobile terminal location of users.The mature location technology has irreplaceable significance.In this paper firstly I summarize the current situation at home and abroad of Base Station Location Technology and related theories.I also introduced advantages and problem of this location technology and related technologies.This laid for the development in the future.Secondly I deeply dissected fingerprint database,and introduced the theories at previous off-line data collection period and the real-time location period.I also analyzed the index variable which was used to recognize fingerprinting database and common process.Finally I simply introduce common process of data mining.Before making mobile terminal positioning,I make selection of original dataset and compiling algorithm which on the basis of network optimizing business requirements.For making further improvement in the efficiency of positioning,I put forward several optimized algorithms which aim at base station location.Then the relative merits and realize process of these algorithms are expounded.And I make result comparison in the end.During the comparison,I found it is successful when we use Bik means clustering algorithm,we will get ideal clustering results of fingerprinting dataset.So the clustering algorithm is adopted in the later positioning,and in the user behavior analysis,the cell business classification is also reflected.The train of thought in this paper is based on the clustering process of data mining acts on fingerprint dataset.Combine with the revised propagation model,I found it worked for more rapid and time saving location service in the future,and also did some groundwork forgetting real-time user data and analyzing user behavior and other researches.